On the basis of our past research on Fe-Mn-Al system, a new Ni-Cr-free cryogenicnon-magnetic steel-15Mn26Al4 has been developed. The steel shows stable austen-ite structure and satisfactory mechanical properties. It may be used for temperaturesas low as that of liquid hydrogen without incurring embrittlement. From our own work on Fe-Mn-Al phase diagram, since aluminum can suppressmartensite type transformation of high manganse austenite steels at low temperaturesand considering that both carbon and manganese promote the stabilization of the austen-ite structure, it is deduced that the chemical composition of the above steel should bein the narrow range of C 0.13-0.19％, Mn 24.5-27.0％ and Al 3.8-4.7％. Such a steelpossesses excellent ductility and impact properties together with higher strength ascompared with that of the conventional 18-8 austenitic stainless steels, although itsrustresisting property is relatively poor. The production of such steel calls for special attention, it is advised to stick to theso-called "3-high and 1-fast" practice, i. e. high smelting temperatures, high slag temper-atures and high mould temperatures with fast teeming. In hot working practice, it hasbeen found that there appeared two regions with high plasticity, 750-850℃ and 1150-1250℃. The 15Mn26Al4 steel has been successfuly fabricated into the sheets, plates, bars,wires, tubes and forgings. At the same time, suitable welding rods coated with properfluxes of different specifications have been developed. Explosion tests to failure at liquid nitrogen temperature of vessels made of 15Mn26Al4 steel were carried out, the broken vessel showed complete plastic dimple fracture,testifying that such a steel is suitable for low temperature uses.

A new process for synthetic rutile by selective chlorination of ilmenite in a chlori-nator with a certain amount of air or oxygen has been investigated. The chlorinatingtemperature and the partial pressure ratio between carbon monoxide and dioxide in thechlorinator are the main factors concerned which are analyzed thermodynamically fromthe point of view of separating iron from titanium. Owing to the fact that the heat evolvedof selective chlorination is insufficient to keep it at temperatures necessary for the reac-tion to be colltinued, a certain excess of carbon was mixed with the ilmenite to provideadditional heat for chlorination. With a view to establish such a new process flow-sheet,operahon condihons including the chlorination temperatures, various ratios of carbonmixed, rates of fiow of oxygen and chlorine, the chlorinating time, particle-sizes of rawmaterials, etc. were examined in laboratory experiments as well as tested on a pilot plantscale and finally checked in industrial practice. The crude rutile obtained was further con-centrated by tabling and magnetic separation. Such a process showed the advantageof being simple, cheap and capable of producing high quality synthetic rutile, besides,the ferric chloride as by-product was also of high-grade.

A high-sulphur-containing coking coal was mixed with alkaline material in order to"fix" its certain amount of sulphur after coking, such coke has been called the "fixing-sulphur" coke. This and another "high-sulphur" coke made from same coal were sepa-rately used in ironmaking under similar conditions for comparison. It was found thatthe "fixing-sulphur" coke showed better properties in ironmaking than the high-sulphurcoke in terms of all technological and economical indices. With "fixing-sulphur" coke,the contamination of sulphur was much less serious, 90％ of pig iron to be casting pro-duced up to the recognized standard. Samples of coke, ore, flux, molten iron and slag were taken from different positionsof blast furnace for sulphur analysis, a diagram of sulphur variation along longitudinaldirection of blast furnace was plotted. Such a diagram may help to control blast furnaceoperation from the point of view of sulphur elimination, although much further investi-gation will be needed in this direction.

The formation and floating-out of slag inclusions in liquid bearing steel have beenstudied with the aid of radioactive isotope Ca~(45) by observing the amount, size, distri-bution, motphology and composition of residual slag inclusions in ingots after solidi-fication. It has been found that the entrapment of slag during tapping could be thecause of the presence globular inclusions in the steel, but that most part of such inclu-sions especially large ones can be eliminated by taking appropriate measures duringmelting and teeming.

A new process for producing medium and low-carbon ferromanganese wherebyhigh-carbon ferromanganese is decarburized by gaseous oxygen in a converter is propos-ed. This paper gives the thermodynamic grounds of the process, in which the temper-ature conditions for the conversion is indicated, the loss of manganese by vaporizationand effect of vacuum estimated.

The steel (18Cr2Ni4WA) melted by three different processes and deoxidized eitherwith aluminium and titanium or with aluminium alone has been studied in order to clar-ify the cause of its rock-like fractography. It is believed that such rock-like fracture isconnected with the sensibility of over-heating of the steel. Conditions render this steelsensible to over-heating are: (1) the over-growth of austenite grains on heating at hightemperatures; (2) the precipitation of fine manganese sulphide inclusions at the austen-ite boundaries. It is considered that small amounts of residual titanium left over afterdeoxidation may retard the growth of austenite grains as well as modify the mode anddistribution of the fine manganese sulphide inclusions, thus reducing the sensibility toover-heating of the steel.

The impact-toughness of a hot-rolled 3％ Si-Fe sheet steel has been measured to in-vestigate its ductile-brittle transition behavior. It was found that the transition from brit-tle to ductile fracture for this steel was within the temperature range of 0 to 100℃.At this range, the impact-toughness value was very sensitive to temperature changes,variation amounted to more than one order in magnitude could happen. Beyond thisrange, however, the curve rised asymptotically to a constant value, at which tempera-ture the sheet steel became ductile. In view of this, preheating the sheet to 60-80℃prior to cold-rolling was found to be an effective way in preventing brittle-cracking,whereas any aging effect which might deteriorate the magnetic properties of the sheetwas found to be neghglble. It is convenient to define the type of deformation after the indices of crystallineplanes being pre-determined by the etch-figure method. Fractography observations bySEM showed that, the {100} planes constituted the majority of the cleavage facets onbrittle fracture, but the narrow transition zone between the different cleavage facetsappeared to be {110} planes. Traces of {110) slip planes on cleavage facets mightoften be found on the cleavage facets, with the main cleavage direction almost parallel tothese slip traces. The majority of larger cleavage steps on the cleavage facets belongedto the {110} planes. During Plastic cracking many high-index crystalline planes otherthan {110) may be observed on the fracture surface, but cleavage along the {100}planes may still occur although with much less frequencies.

The behaviour of rare earth elements, taking cerium as representative, in moltencommercial pure iron and their reactions on refractories have been inveshgated bymeans of autoradiography and radioassay. Results show that under steelmaking tem-peratures the evaporation of cerium from molten iron or cerium oxide from slag isvery slight, and cerium in molten iron reacts in different degrees on crucibles made offireclay, high-alumina, magnesia, alumina-magnesia, silica, as well as alundum, electro-fused-magnesia and zirconia. These erosive reactions are affected not only by the chem-ical constituents but also by the physical properties of refractories. It was observed byautoradiography that the erosive reaction products may spall away then to form rareearth inclusions in iron consequently. A considerable part of these reaction productsand the rare earth inclusions do not float to the surface of liquid metal, and may adhereto the crucible wall. The blockage on laddle nozzle and runner during teeming has beenalso discussed.

The present paper deals with the method of locating the crack initiation point inJ_(IC) determinahon by means of a single specimen. The first is an analytical method forcalculating crack length variations during monotonic loading from the load-displace-ment record alone, and the second is based on And-sectional measurements of C. T. O. D.Results of application of these methods are presented which demonstrate the effective-ness of the methods proposed. In additional, some recent experimental results are also reported which substanti-ate the slip-line field proposed earlier for the CT specimen, the result for the limit loadP_L, and the J-U relation obtained by us.